1
Robert PittDepartment of Civil, Construction, and Environmental
EngineeringUniversity of Alabama
Tuscaloosa, AL, USA 35487
The Selection of an Urban Runoff The Selection of an Urban Runoff Control Program using Decision Control Program using Decision
AnalysisAnalysis
Photo by Lovena, Harrisburg, PA
Stormwater Effects• Sediment (amount and quality)• Habitat destruction (mostly through high flows
[energy] and sedimentation)• Eutrophication (nutrient enrichment)• Low dissolved oxygen (from organic materials)• Pathogens (urban wildlife vs. municipal wastewater)• Toxicants (heavy metals and organic toxicants)• Temperature• Debris and unsafe conditions• etc.
Probability distribution of rains (by count) and runoff (by depth).
Birmingham Rains:<0.5”: 65% of rains(10% of runoff)
0.5 to 3”: 30% of rains(75% of runoff)
3 to 8”: 4% of rains(13% of runoff)
>8”: <0.1% of rains(2% of runoff)
0.5” 3” 8”
Birmingham, AL, rains from 1952 through 1989
111 rains per year during this 37 year periodMost rains < 3 inchesAbout 5 rains a year between 3 and 8 inches3 rains (in 37 years) > 8 inches
2
Suitable Controls for Almost Complete Elimination of Runoff Associated with
Small Rains (<0.5 in.)
• Disconnect roofs and pavement from impervious drainages
• Grass swales• Porous pavement walkways• Rain barrels and cisterns for local
reuse
Suitable Controls for Treatment of Runoff from Intermediate-
Sized Rains (0.5 to 3 in.)• Initial portions of these rains will be
captured/infiltrated by on-site controls or grass swales, but seldom can infiltrate all of these rains
• Remaining portion of runoff should be treated to remove particulate-bound pollutants
Roof drain disconnections
Not this!
Rain Garden Designed for Complete Infiltration of Roof Runoff
3
Land and Water, Sept/Oct. 2004
97% Runoff Volume Reduction
Calculated Benefits of Various Roof Runoff Controls (compared to typical directly connected residential pitched roofs)
100
87
77
67
21
Seattle, Wash. (33.4 in.)
87
84
75
66
13
Birmingham, Alabama (55.5 in.)
96%Rain garden with amended soils (10 ft. x 6.5 ft.)
91%Disconnect roof drains to loam soils
84%Planted green roof (but will need to irrigate during dry periods)
88%Cistern for reuse of runoff for toilet flushing and irrigation (10 ft. diameter x 5 ft. high)
25%Flat roofs instead of pitched roofs
Phoenix, Arizona (9.6 in.)
Annual roof runoff volume reductions
Runoff from Pervious/
impervious area
Trapping of sedimentsand associated pollutantsReducing velocity of
runoff
Infiltration
Reduced volume and treated runoff
Sedimentparticles
Particulate Removal in Shallow Flowing Grass Swales and in Grass Filters
Head (0ft)
Date: 10/11/2004
2 ft
25 ft
6 ft
3 ft
116 ft75 ft
TSS: 10 mg/L
TSS: 20 mg/L
TSS: 30 mg/L
TSS: 35 mg/L
TSS: 63 mg/L
TSS: 84 mg/L
TSS: 102 mg/L
University of Alabama swale test site at Tuscaloosa City Hall
4
WI DNR photo
Conventional curbs with inlets directed to site swales Grass Swales Designed to Infiltrate Large Fractions of Runoff (Alabama).
Also incorporate grass filtering before infiltration
Porous paver blocks have been used in many locations to reduce runoff to combined systems, reducing overflow frequency and volumes (Sweden, Germany, and WI).
Not recommended in areas of heavy automobile use due to groundwater contamination (provide little capture of critical pollutants, plus most manufactures recommend use of heavy salt applications instead of sand for ice control).
Bioretention and biofiltration areas having moderate capacity
5
Recent Bioretention Retrofit Projects in Commercial and Residential Areas in Madison, WI
Soil Compaction and Recovery of Infiltration Rates
• Typical site development dramatically alters soil density.
• This significantly reduces infiltration rates, especially if clays are present.
• Also hinders plant growth by reducing root penetration
Long-Term Sustainable Average Infiltration Rates
0.290.015<0.001
1.5021.7031.911
HandStandardModified
Clay Loam
1.30.0270.0017
1.5041.5931.690
HandStandardModified
Silt Loam
3591.5
1.5951.6531.992
HandStandardModified
Sandy Loam
Long-term Average Infilt. Rate (in/hr)
Dry Bulk Density (g/cc)
Compaction Method
SoilTexture
Compaction, especially when a small amount of clay is present, causes a large loss in infiltration capacity.
6
Soil Modifications for rain gardens and other biofiltration areas can significantly increase treatment and infiltration capacity compared to native soils.
Rob Harrison ,Univ. of Wash., and Bob Pitt, Univ. of Alabama examined the benefits of adding large amounts of compost to glacial till soils at the time of land development (4” of compost for 8” of soil)
3.3UW test plot 6 Alderwood soil with GroCocompost (old site)
0.3UW test plot 5 Alderwood soil alone
3.0UW test plot 2 Alderwood soil with Ceder Grovecompost (old site)
0.5UW test plot 1 Alderwood soil alone
Average Infiltration Rate (in/h)
Enhanced Infiltration with Amendments
Six to eleven times increased infiltration rates using compost-amended soils measured during long-term tests using large test plots and actual rains (these plots were 3 years old).
Changes in Mass Discharges for Plots having Amended Soil Compared to Unamended Soil
0.180.061Zinc
1.20.33Copper
1.50.28Nitrate
4.40.56Ammonia
3.00.62Phosphate
0.29 (due to ET)0.09Runoff Volume
Subsurface Flow Mass Discharges
Surface Runoff Mass Discharges
Constituent
Increased mass discharges in subsurface water pollutants observed for many constituents (new plots).
Tests on Soil Amendments
• Many tests have been conducted to investigate filtration/ion exchange/sorption properties of materials that can be potentially used as a soil amendment.
7
Some laboratory and field pilot-scale test setups (EPA and WERF-supported research at Univ. of Alabama). Critical that tests use actual stormwater, not artificial mixtures.
Capture of Stormwater Particulates by Different Soils, Media, and Amendments
100%100%100%100%80%45%40%Activated carbon, peat, and fine sand
100%50%25%0%0%0%0%Loam soil
100%100%100%90%85%33%10%Fine sand
10%0%0%0%0%0%0%Coarse gravel
100%50%25%10%0%0%0%Porous pavement surface (asphalt or concrete)
>250µm120 to 250µm
60 to 120µm
30 to 60µm
12 to 30µm
3 to 12µm
0.45 to 3µm
Measured Particle Sizes, Including Bed Load Component, at Monroe St. Detention Pond, Madison, WI
8
COD and phosphorus concentrations vary as a function of particle size
Laboratory Media Studies • Rate and Extent of Metals Capture– Capacities
(partitioning)– Kinetics (rate of
uptake)
• Effect of pH & pH changes due to media, particle size, interfering ions, etc
• Packed bed filter studies
• Physical properties and surface area determinations
Cation Exchange Capacities for Different Media
3.5Sand
9.4Agrofiber
3.8Cotton Waste
6.9Zeolite
5.4Activated Carbon
19Compost
22Peat Moss
CEC (meq/100 g)
Sand had very little capacity
Compost leached soluble P during all conditions, especially if anaerobic
Peat had greatest capacity for P
Contaminant Losses during Anaerobic vs. Aerobic Conditions between Events
9
Unfortunately, peat harvesting is a surface mining operation of a non-renewable resource.
Locally available organic wastes, appropriately processed, should be investigated as a preferable soil amendment.
Near Tullamore, County Offaly, Ireland
Conservation Design Approach for New Development
• Better site planning to maximize resources of site• Emphasize water conservation and water reuse on
site• Encourage infiltration of runoff at site but prevent
groundwater contamination• Treat water at critical source areas and encourage
pollution prevention (no zinc coatings and copper, for example)
• Treat runoff that cannot be infiltrated at site
Conventional Development
Conservation Design
Sediment Reductions
Volume Reductions
10
Stormwater Infiltration Controls in Urban Areas
• Bioretention areas• Rain gardens • Porous pavement• Grass swales • Infiltration Basins• Infiltration Trenches• Subsurface Dispersal
Seattle, WA
Wet Detention Ponds
Suspended Solids Control at Monroe St. Detention Pond, Madison, WI (USGS and WI DNR data)
Consistently high TSS removals for all influent concentrations (but better at higher concentrations, as expected)
Critical Source Area ControlCovering fueling area
Berm around storage tanks
11
Multi-Chambered Treatment Train (MCTT) for stormwater control at large critical source areas
Milwaukee, WI, Ruby Garage Maintenance Yard MCTT
Pilot-Scale MCTT Test
Results
Ruby Garage MCTT samplesinfluent effluent
EPA-funded SBIR2 Field Test Site Monitoring Equipment, Tuscaloosa, AL
Upflow filter insert for catchbasins
Able to remove particulates and targeted pollutants at small critical source areas. Also traps coarse material and floatables in sump and away from flow path.
Pelletized Peat, Activated Carbon, and Fine Sand
y = 2.0238x0.8516
R2 = 0.9714
0
5
10
15
20
25
0 5 10 15 20
Headloss (inches)
Flow
(gpm
)Performance Plot for Mixed Media on Suspended Soilds for Influent
Concentrations of 500 mg/L, 250mg/L, 100 mg/L and 50 mg/L
0
100
200
300
400
500
600
Influent Conc. Effluent Conc.
Susp
ende
d So
ilds
(mg/
L)
High Flow 500Mid Flow 500
Low Flow 500
High Flow 250Mid Flow 250
Low Flow 250
High Flow 100Mid Flow 100
Low Flow 100High Flow 50
Mid Flow 50Low Flow 50
HydroInternational, Ltd.
12
Benthic macroinvertebrate populations on natural and artificial substrates have been extensively used to indicate receiving water effects.
EXCELLENT
GOOD
FAIR
POOR
0
10
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90 100Watershed Urbanization (%TIA)
0
5
10
15
20
25
30
35
40
45
Ben
thic
Inde
x of
Bio
tic In
tegr
ity
(B-I
BI)
Riparian IntegrityBiotic Integrity
C. May 1996
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1 10 100
Directly Connected Imperv Area (%)
Rv
(san
dy s
oils
)
WinSLAMM predicts biological conditions in receiving waters based on reduced runoff quantities and effective impervious areas
00.10.20.30.40.50.60.70.80.9
1
1 10 100Directly Connected Impervious Area (%)
Rv
Sandy Soil Rv Silty Soil Rv Clayey Soil Rv
GoodFair
Poor
Relationship between Directly Connected Impervious Areas, Volumetric Runoff
Coefficient, and Expected Biological Conditions
WinSLAMM calculates flow-duration curves for site, with and without controls
13
Flow-Duration Curves for Different Stormwater Conservation Design Practices
0
20
40
60
80
100
120
140
0.1 1 10 100
% Greater than Discharge Rate
Dis
char
ge (c
fs)
Top Set:No ControlsSwales
Bottom Set:BiorententionSwales and BioretentionPond and Bioretention Pond, Swales and Bioretention
Flow Duration Curves are Ranked in Order of Peak Flows
Middle Set:PondPond and Swales
Capture and Reuse of Roof Runoff for Supplemental Irrigation
9816,000
908,000
744,000
562,000
56%1,000
Percentage of Annual Roof Runoff used for Irrigation
Tankage Volume (ft3) per 4,000 ft2 Building
Stormwater Control Categories in the International Stormwater “BMP” Database:
Structural Controls:•Detention ponds•Grass filter strips•Infiltration basins•Media filters•Porous pavement•Retention ponds•Percolation trenches/wells•Wetland basins•Wetland channels/swales•Hydrodynamic devices
Non-Structural Controls:•Education practice•Recycling practice•Maintenance practice•Source controls
Decision Analysis
• With so much data available, and so many options that can be analyzed, how does one select the “best” stormwater control program?
• The least costly that meets the objective?
14
Possible, if only have one numeric standard:
If 80% SS reduction goal, the least costly would be wet detention. In this example, grass swales, street cleaning, and catchbasins cannot reach this level of control. If 40% SS reduction goal, then grass swales wins.
But, recall that there are many elements that must be considered:
• Sediment (amount and quality) habitat destruction (mostly through high flows [energy] and sedimentation)
• Eutrophication (nutrient enrichment)• Low dissolved oxygen (from organic materials)• Pathogens (urban wildlife vs. municipal wastewater)• Toxicants (heavy metals and organic toxicants)• Temperature• Debris and unsafe conditions• etc.
A multi-attribute decision analysis procedure can be used to examine many conflicting objectives. One example is by Keeney and Raiffa (Decision Analysis with Multiple Conflicting Objectives). This method uses utility curves to describe the benefits of varying levels of control and tradeoff coefficients that compare the different objectives. The first step is to determine the outcomes for several alternative stormwater control programs that will be compared. WinSLAMM can assist with this using the batch processor:
This is an example WinSLAMM batch processor output, showing many features (including costs, performance, habitat effects, etc.) for eight alternative programs:
15
Sum = 1.0
0.125.5 to 25 lbs/yPart. Phosphorus yield (lbs/y)
0.072,183 to 10,192 lbs/yParticulate solids yield (lbs/y)
0.180 to 0.05 %% of time flow >10 cfs0.050.5 to 4 %% of time flow >1 cfs0.300.06 to 0.29Rv 0.082.3 to 4.5 acresLand needs (acres)0.20$40,217 to 83,364Total annual cost ($/year)
Trade-offs between
remaining attributes
Range of attribute value for acceptable options
Attribute
Example ranges of attributes, and trade-offs:
• Total annual cost: straight line, with $83,364 = 0 and $40,217 = 1.0.
Example Utility Values for Attributes:
Utility Function for Annual Costs
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 20000 40000 60000 80000 100000
Annual Costs ($/year)
Util
ity V
alue
% of time flow >10 cfs Utility value
<0.05 1.00.05 - 1 0.751.1 – 2.5 0.25>2.5 0
Utility Curve for Flows >10 CFS
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.5 1 1.5 2
% of time flow > 10 CFS
Util
ity V
alue
• Part. Phosphorus yield (lbs/yr): straight line, with 25 lbs/yr = 0 and 5.5 lbs/yr = 1.0
• Land needs (acres): straight line, with 4.5 acres = 0 and 2.3 acres = 1.0
• Particulate solids yield (lbs/yr): straight line, with 10,192 lbs/yr = 0 and 2,183 lbs/yr = 1.0
• % of time flow >1 cfs Utility value<1 1.01 – 3 0.753.1 – 10 0.25
>10 0
Example Utility Values for Other Attributes (cont):
16
0.77100.764,12512.30.8745,698Option 8Small pond, reg. swale and biofilter
0.41170.416,93712.3140,217Option 7Small pond and reg. swale
1.05.51.02,18304.50.6754,622Option 6Pond, reg. swale and biofilter
0.77100.764,13304.50.7949,142Option 5Pond and reg. swale
025010,19204.5083,364Option 1Pond
0.120.070.080.20Tradeoff Value
Phos. utility
Part. Phos. Yield
(lbs/yr)
Part. Solids utility
Part. Solids Yield
(lbs/yr)
Land utility
Land Needs
for SW mgt
(acres)
Cost utility
Total Annual
Cost ($/yr)
Stormwater Control Option
Attribute Values and Associated Utilities for Example
10.92900.181.00.051.00.301.0Option 8Small pond, reg. swale and biofilter
30.75550.1350.750.03750.750.2250.75Option 7Small pond and reg. swale
20.85400.181.00.051.00.301.0Option 6Pond, reg. swale and biofilter
4 0.74550.181.00.03750.750.2250.75Option 5Pond and reg. swale
50.22250.1350.750.01250.250.0750.25Option 1Pond
0.180.050.30Tradeoff Value
Over-all
Rank
Sum of factors
High flow
factor
High flow
utility
Mod flow
factor
Mod flow
utility
Rv factorRv utility
Stormwater Control Option
Calculation of Factors for Each Option (cont.), Sum of Factors, and Overall Rank
Appropriate Combinations of Controls• No single control is adequate for all problems• Only infiltration reduces water flows, along with soluble
and particulate pollutants. Only applicable in conditions having minimal groundwater contamination potential.
• Wet detention ponds reduce particulate pollutants and may help control dry weather flows. They do not consistently reduce concentrations of soluble pollutants, nor do they generally solve regional drainage and flooding problems.
• A combination of bioretention and sedimentation practices is usually needed, at both critical source areas and at critical outfalls.
Pitt, et al. (2000)
• Smallest storms should be captured on-site for reuse, or infiltrated
• Design controls to treat runoff that cannot be infiltrated on site
• Provide controls to reduce energy of large events that would otherwise affect habitat
• Provide conventional flood and drainage controls
Combinations of Controls Needed to Meet Many Stormwater Management Objectives
17
AcknowledgementsA large number of projects were briefly represented during this presentation. This research has been supported by many, including:• The US Environmental Protection Agency (EPA)• The National Science Foundation (NSF)• Water Environment Research Foundation (WERF)• Auckland Regional Council, New Zealand• CEDARS, the Center for Economic Development and Resource Stewardship and TVA• The Stormwater Management Authority of Jefferson County (SWMA), City of Tuscaloosa, AL DOT, the University of Alabama• States of Wisconsin, New York, California, Washington, etc. • GeoSyntec, HydroInternational, Earth Tech, etc.• Of course, many graduate students also contributed greatly.
The opinions expressed during this presentation are the authors and do not necessarily reflect these sponsors!